本文首先將搜集發光二極體(LED)檯燈的繁體和簡體中文及英文專利文件，並進行LED檯燈繁體和簡體中文及英文專利的IPC分析。此外本文亦發展繁體和簡體中文及英文專利之創新性零組件元件字斷詞法，及擴充專利相關技術詞與功能詞之第一層斷詞點，進一步建立創新第二層斷詞點的方法。依照前述之繁體和簡體中文及英文斷詞斷字法方法，本文歸納繁體和簡體中文及英文專利之零組件關鍵字及技術詞句與功能詞句。並應用所搜集的LED檯燈相關繁體和簡體中文及英文專利，收集同義字，建立繁體和簡體中文及英文同義字對照表。接著建立LED檯燈相關繁體和簡體中文及英文專利的技術/功能矩陣，並進行以技術觀點為主的LED檯燈繁體和簡體中文及英文專利分析。本文並應用判斷一件專利所歸屬的技術或功能類別之方法。此方法為先利用上述建立的繁體和簡體中文及英文同義字表，配合元件字、技術字群及功能字群之關鍵字之出現次數，建立關鍵字相關的模糊值，進而應用次數常態化概念，建立相關公式，算出該繁體和簡體中文及英文專利可能歸屬之技術或功能類別的概率值。
接著本文以LED檯燈為載具，進行發明人及國別之各類IPC統計圖表分析。並將建立以公司為欄，IPC碼為列，專利篇數為元素的競爭者核心競爭力圖﹔以及公司為欄，年份為列，專利篇數為元素的的主要競爭者研發活動趨勢圖。又依上述所建立的創新繁體和簡體中文及英文專利斷詞和斷字方法，進行LED檯燈的繁體和簡體中文及英文專利之元件字、技術詞和功能詞的相關斷詞及斷字分析，再半自動半人工進行前述相關繁體和簡體中文及英文專利之技術及功能分析。進而建構LED檯燈相關繁體和簡體中文及英文專利之技術/功能矩陣。應用所建立之技術/功能矩陣，本文進而建立以技術為主之發明人及國別的各類統計圖表分析，以及競爭者核心競爭力圖。進而將技術為主之分析結果與IPC分析法之結果相比較。上述所搜尋的相關LED檯燈的繁體和簡體中文及英文專利，以及前述之IPC分析和技術分析之結果，可提供給企業及工程師作為參考，以節省研發時間，作為新專利產生之參考，此外亦可使企業了解競爭者之核心競爭力。

First of all, the paper collects the traditional Chinese, simplified Chinese and English patent documents of light emitting diode (LED) table lamp, and conducts international patent classification (IPC) analysis of the Chinese and English patents of LED table lamp. Besides, the paper also develops an innovative term segmentation and word segmentation method of the components of Chinese and English patents, expands the first layer’s term segmentation points of patent-related technical terms and functional terms, and further establishes an innovative method of the second layer’s term segmentation points. According to the abovementioned term segmentation and word segmentation method, the paper induces the keywords, technical terms and functional terms of the components of Chinese patents and English patents, and also applies the established table of traditional Chinese, simplified Chinese and English patents of LED table lamp to collect synonyms. After that, the paper establishes the technical/function matrix of the Chinese and English patents of LED table lamp, and makes technical-viewpoint-based analysis of traditional Chinese, simplified Chinese and English patents of LED table lamp. If it is required to judge what technical or functional kind a patent belongs to, the paper uses the table of Chinese and English synonyms established above and the number of occurrence times of keywords to establish the related fuzzy values. In addition, the paper applies normalization concept of the number of times to establish the related equations, and further calculates the probability Pj that a patent may belong to a certain functional or functional kind. After that, with LED table lamp taken as an important carrier, the paper analyzes the various IPC statistical charts of inventors and countries. Besides, the paper establishes a table of competitors’ core competitiveness, with company taken as the column, IPC code as the row, and the number of patents as the element; as well as a trend chart of main competitors’ research and development (R&D) activities, with company taken as the column, year as the row, and the number of patents as the element. According to the innovative term segmentation and word segmentation method of the traditional Chinese, simplified Chinese and English patents established above, the paper makes the related term segmentation and word segmentation analysis on the component words, technical terms and functional terms of the traditional Chinese, simplified Chinese and English patents of LED table lamp. After that, the paper semi-automatically and semi-manually makes technical and functional analysis of the abovementioned patents, and further constructs a technical/functional matrix of the related patents of LED table lamp. Using the established technical/functional matrix, the paper further establishes different statistical charts of technique-based inventors and countries for analysis, and a chart of competitors’ core competitiveness. Furthermore, the charts are compared with the results achieved by IPC analytical method. The related traditional Chinese, simplified Chinese and English patents of LED table lamp searched above, as well as the results acquired from the above IPC analysis and technical analysis can provide useful references for enterprises and engineers in producing new patents and save their R&D time, and also can make enterprises understand the core competitiveness of competitors.

[1] Lawrence, S., and Giles, C. L., (1999), Searching the World Wilde Web General and Scientific Information Access, IEEE Communication, 37(1):116-122.
[2] Trappey, C. V., Trappey, A. J. C., and Wu, C. Y., (2010), Clustering patents using non-exhaustive overlaps”, Journal of Systems Science and Systems Engineering, 19(2):162-181.
[3] Trappey, C. V., Trappey, A. J. C., and Wu, C. Y., (2009), Automatic patent document summarization for collaborative knowledge systems and services, Journal of Systems Science and Systems Engineering, 18(1):71-94.
[4] Singhal, A. and G. Salton, “Automatic Text Browsing Using Vector Space Model, ”Technical Report, Department of Computer Science, Cornell University, (1993).
[5] Gruber, T., “Ontolingua: A translation approach to portable ontology specifications,” Knowledge Acquisition, pp.199-200,(1993).
[6] Oleary, D. E,“Enterprise knowledge management,” Computer IEEE, Vol.31, pp.54-61 (1998).
[7] 拓墣產業研究所，2007深入剖析LED產業市場發展，拓墣科技，台北市(2007)
[8] 陳哲宏、陳逸南、謝銘洋、徐宏昇，專利法解讀，月旦出版公司，第14-15頁，民83年。
[9] Ryan J. M., Roshan J. and Song L., “Integrated Thermal Management Techniques for High Power Electronic Devices”, Applied Thermal Engineering, Vol.24, Issue 8-9, pp.1143-1156 (2004).
[10] 「國際專利分類表第七版使用指南」，經濟部智慧財產局，民90年，7月。
[11] 「國際專利分類之研究」，經濟部中央標準局。
[12] Overview of the Classification System, http://www.uspto.gov/main/patents.htm
[13] The EPO guide to patent information on the internet, http://epart.epo.org/dwl/espacenet_manual.pdf
[14] 謝寶煖，「專利與專利資訊檢索」，大學圖書館，2卷4期，第111-127頁，民87年，10月。
[15] 賴奎魁、張瑩珠、鄭伶如，建立專利策略群組模式之研究，第853-867頁，1999全國智慧財產權研討會，國立交通大學企業法律中心，民88年。